PROJECT SUMMARY/ABSTRACT The control of membrane shape and curvature is essential for many vital cellular functions, including cell division and cell motility, as well as vesicle budding and fusion. Recent work has shown that these processes are regulated by proteins that can sense, stabilize or induce membrane curvature. Not surprisingly, aberrations in the control of membrane curvature have been linked to a number of diseases, including mental retardation, cancer and muscular dystrophy. The underlying molecular mechanisms, however, are poorly understood since little structural information exists for the biologically relevant membrane-bound forms of the curvature-sensing or inducing proteins. The central goal of this proposal is to provide such detailed structural and mechanistic information. Aim 1 investigates the molecular mechanisms by which annexins can sense membrane curvature. This remarkable process causes a major inside-out refolding of annexins and brings buried hydrophobic residues of the solution structure into direct contact with the acyl chains of the membranes. Using a combination of continuous wave and pulsed EPR spectroscopy, the proposed work will define the three-dimensional structure of this membrane-bound form and test the hypothesis that N-terminal phosphorylation modulates curvature- dependent membrane interaction in vivo. Furthermore, it will be tested, whether the structure of the curvature- dependent membrane bound forms is related to that of the interfacial, pH-dependent membrane-bound form. Aim 2 investigates the membrane-bound forms of the curvature-inducing N-BAR proteins endophilin and amphiphysin, while aim 3 addresses the structure of the membrane-bound form of the F-BAR protein FCHo2. The detailed structural studies proposed in both aims are designed to test the hypothesis that BAR domain proteins induce membrane curvature using a combination of three different mechanisms: (1) by acting as scaffolds that mold a specific membrane curvature; (2) by wedging amphipathic helices into the membrane; and (3) by forming specifically aligned oligomeric structures. In particular it will be tested whether N-BAR and F-BAR proteins induce curvature by using the aforementioned mechanisms to different extents. All structural data obtained from the EPR analysis will be refined further using the recently developed PRONOX algorithm which will be employed to generate three-dimensional structural models that contain the proteins as well as the lipid membranes. These models will provide direct insight into the molecular mechanisms by which BAR domain-containing proteins induce membrane curvature. Structural analysis will further be used to test the mechanism by which amphiphysin-2 mutants (K35N and D151N) have reduced curvature-inducing properties and misfunction in familial forms of centronuclear myopathy.